Intraoperative consultation/final diagnosis correlation: relationship to tissue type and pathologic process.

The study analyzed all intraoperative consultation (IC) cases from Calgary Laboratory Services over a 12-month period. A single pathologist reviewed each case to extract relevant data points. The dataset included intraoperative and final diagnoses, specimen types, and diagnostic intent. The researchers calculated overall agreement rates and used statistical measures like the kappa statistic. They categorized discrepancies into interpretive and sampling errors. Specimen types were grouped into categories such as lymph nodes, thyroid, and nervous system tissues. The analysis also tracked deferrals and disagreement types. The study focused on identifying patterns of diagnostic accuracy by tissue type and diagnostic purpose. This approach allowed for a detailed evaluation of IC performance across different clinical scenarios. The researchers ensured data consistency by using standardized criteria for agreement and disagreement.

The study found a high overall agreement rate of 96.75% between intraoperative consultations (IC) and final diagnoses. The kappa statistic of 0.94 indicated strong inter-rater reliability. Out of 2812 specimens, 87 showed discordance and 135 were deferred. Lymph nodes, thyroid/parathyroid, and nervous system tissues were the most frequently submitted for IC. These three tissue types also accounted for the majority of disagreements. Neoplasm typing and margin assessment were the most common diagnostic requests and sources of discordance. Interpretive errors were the leading cause of disagreement, followed by gross sampling errors. False-negative results occurred nearly three times more often than false positives. These findings highlight the diagnostic challenges in specific tissue types and assessment categories.

The study confirmed that intraoperative consultation (IC) is a highly accurate diagnostic tool in surgical pathology. The high agreement rate of 96.75% supports its role in guiding intraoperative decision-making. However, certain tissue types and diagnostic purposes showed higher rates of discordance. Lymph nodes, thyroid, and nervous system specimens accounted for the most disagreements. Neoplasm typing and margin assessment were the most frequently requested assessments and also the most error-prone. Interpretive and sampling errors were the primary causes of diagnostic discrepancies. False-negative results were more common than false positives, suggesting a need for caution in these cases. These findings suggest that diagnostic accuracy varies depending on specimen type and diagnostic intent. The authors propose that these insights can inform quality assurance protocols in surgical pathology.